Simultaneous 3 compartment model plasma plus tissue

Phoenix Certara Forums Modelling and Simulation
1

Hi,

I have plasma as well as tissue concentration (Liver and brain) and I would like to develop 3 compartment model for the drug using all three concentration.

Could anyone tell how to do in Phoenix NLME? How to prepare a data file and how to add 3 observation in the model? It would be great if someone shares the datafile preparation and show how to map in NLME.

2

Morning, This should be fairly straightforward, at least in model definition. I think it would be easier to have the measurements for each compartment as their own column of data, e,g, c_plasma, c_brain, c_liver. Choose a base model e.g. 2 compartments and then once you have chosen “edit graphical” you can right click in your model to add extra observations and compartments as required.

I would look at the NLME examples guide, to become more familiar with the model building e.g. Graphically create a peripheral elimination model (page 177 in v8.1 documentation)

and also look at some of the recorded PML school webinars e.g. https://support.certara.com/forums/topic/1116-minimal-physiologically-based-pharmacokinetic-model-for-monoclonal-antibodies.

Try and that and if you get stuck post the project file back here and we can try to advise further.

Simon.

3

keep in mind that the compartmental models do not have physiological meaning except very few unique cases.

If you are really looking for a model representing Brain and Heart you need to go into PBPK and PML can be coded that way you can make your PBPK a simplified one depending on your needs.

As Simon mentioned you can have sperate columns or your can have one column with a differentiator between the observations

at a given time point you might have one type of observation only and that is fine you can leave cells empty Phoenix understand empty cells.

4

Thanks for your reply. I was able to fit my model.

5

Hi,

If we use a compartmental model to describe plasma (mg/mL) and tissue concentrations (mg/g), how do we estimate volume of distribution of the tissue compartments?

We use k12V1=k21V2 to derive V2 in a typical two-compartment model. In this case, V1 is central Vd in mL and V2 is “tissue compartment volume” and in a different unit, g. Can we simply add V2 as a parameter to be estimated? That would seem to make the estimation of k12 and k21 more difficult.

Thanks, any advice or help is appreciated.

6

you may find parametising in clearance rather than micro constants gives more stable solutions.